Inverse kinematics solution for robot manipulator based on adaptive MIMO neural network model optimized by hybrid differential evolution algorithm

In this paper, a new hybrid differential evolution algorithm is proposed, which combines the differential evolution (DE) algorithm and the back-propagation (BP) algorithm. This new hybrid algorithm is used to train an adaptive MIMO neural network (or AMNN) model for identifying the inverse kinematics of the industrial robot manipulator. Simulation results prove that the proposed identification process of the new hybrid algorithm performs faster convergence and better precision than the conventional back-propagation algorithm or the solely differential evolution algorithm. Consequently, the inverse kinematics of the industrial robot manipulator identification based on the AMNM achieves outstanding performance.